How Much Is Enough? First Steps to a Social Ecology of the Pergamon Microregion

Round 1

Reviewer 1 Report

Dear authors,

The manuscript entitled "How much is enough? First steps to a social ecology of the Pergamon micro-region" provides an original and innovative methodological approach to the environmental carrying capacity in Roman times. It consists of a socio-ecological approach of much relevance for archaeological studies, methodology, results and the provided discussion match perfectly in a really interesting modelling application. I recommend its publication.

Author Response

Dear reviewer,

we are very happy that you find our study insightful and the modelling approach viable and fruitful.

Best wishes,
Julian Laabs & Daniel Knitter

Reviewer 2 Report

Julian Laabs and Daniel Knitter propose a combination of a theoretical/methodological approach of the so-called central place theory and environmental modelling to understand the center-hinterland functionalities during the Hellenistic/Roman period in Asia Minor with a particular emphasize of the catchment/complementary area of Pergamon. The main goal of the authors is to evaluate potential land-use and agricultural resource exploitation, which enabled the local to regional food supply of the central place. They raise the question, to which extent the economic function of the center/hinterland interaction was limited by available and suitable cropland or by demographic numbers and availability of labour. In this context, the authors do not focus primarily on the theoretical discourse based on Christaller’s work from the 1930’s but rather take a presumed center (Pergamon) as initial point to develop three different hinterland/catchment/complementary regions, which derive from three different heuristics. It is important to notice that the authors decided to propose ‘heuristics’ instead of hypothesis or a theoretical model, which allows to emphasize the incomplete knowledge of the archaeological record, the fuzzy approach of the model, and the resolution/determination of data, which was chosen to be part of the model. With this approach, the authors further emphasize the ‘first step’ research (see title) and the potential of the model to be further developed by future research.

From the three heuristics, three different complementary regions were calculated, from which the center of Pergamon could have been supplied with goods (in this case only cereal crops and leguminosae).  A) a core territory, the reconstructed chora, B) a so-called Pergamon Micro-Region, which includes territorial aspects of other villages and towns, and C) a catchment based on hydrological features and the main river courses of the region. Figure 1 describes theses features, however, it would have been nice to include more topographical/hydrological information in the map (e.g. villages that are mentioned in the text, rivers, which determine the catchment etc.). I am sure this is easy to include in the revision.

In the introduction, the authors provide a lot of information. The methodical approach is outlined, followed by geographical settings of the region and basic archaeological development. After that (line 78) there is another section, which covers geographical/environmental information. Maybe also due to the placement of Fig.1, this part appears to be somewhat disconnected from the upper part. Maybe consider restructuring here or, which I personally prefer, insert a brief ‘environmental settings’ section to increase readability.

In the methods section, the authors correctly highlight the heuristic approach of (archaeological) research, which aims at tracing past land-use and/or demographic development. That is an important issue and I want to emphasize the author’s awareness of the problem (which sometimes is all to rapidly glossed over).

In the following, Laabs and Knitter define the concept of Carrying Capacity, including Malthusian theory, and the way they (among other authors) determine land suitability. Instead of absolute data, the authors use a fuzzy logic approach, which builds on previous work by Daniel Knitter and colleagues. The suitability maps proposed by the authors follow a rather simple consideration of topographic elements (i.e. slope gradient) and TWI (Topo. wetness index), which enables the determination of soil moisture on a very basic level and on generalized topographic features (derived from a digital elevation model, DEM, here generalized to a 50 m grid resolution). As the authors state, these model-parameters provide a simple and reproducible tool and need further evaluation by including other environmental components, e.g. soil units, climate variability, geological data, drainage potential etc. The authors show that they are aware of these limitations and that their model (in the sense of a model as a simplification of a potentially realistic approach) works with the selected parameters (so it is consistent in itself). That’s an important statement because I am sure that critics can be raised concerning the implementation only of selected environmental data… but that is a common critic in digital archaeology and this paper does well in pointing this out!

Another parameter in the author’s land-use model is the amount of labour, which could have been taken into account by variable population size, and dietary habits and calory needs of ancient populations. That is an interesting part of the model, which, unfortunately!, cannot be evaluated in detail in this review due to the fact that the important table 2 was missing in the pdf-Manuscript…

In the text, the authors provide some information, which could also be visualized in the table, so I do my best to get the most important parts from that. From one paper (Teegen, 2017), the authors conclude that the ‘average’ Pergamean individual’s diet was plant-based. Right now, I don’t have the paper at hand and because I cannot see the data, I need to rely on the author’s statements. However, isotope data interpretation is tricky, as has been pointed out by a great many authors, and relies greatly on the (micro-)regional, local, and site-specific baseline (see recently Depaermentier et al. 2020, Tracing mobility … Plos ONE). In this section, the authors focus mostly on Roman dietary habits and they conclude that 81% of the average diet is composed of cereals. They further propose 14% to be composed of legmuninosae and another 15% of animal-derived dietary products e.g. cheese, milk and others like wine, nuts etc. There must be something wrong with the numbers here, so please revise this section.

In the following sections, the authors propose the calculation, which underlies the modelling approach. These sections are generally well-written and provide – as stated in the introduction – a reproducible model for future analyses. However, at line 258, there starts a little discussion about Roman labor input during harvest and generally agricultural crop production and I would like to see more references there, which consider different approaches to the amount of land a single person could manage during a specific time period. This is strongly dependent on the terrain, soil, and plant-characteristics and despite the fact that I am aware of the model-character and the generalization needed, a more critical review of literature could enhance the significance of the presented approach.

The results and discussion sections start with an informative figure, which visualizes the land-suitability under the assumption of low and high labour input for the three heuristic complementary regions. From the computed data, the authors determine a threshold of 0.5 to be decisive for the categorization of ‘suitable land’. That is tricky, because the author’s propose a fuzzy approach (low suitability, high suitability) and the combination of a given threshold can be seen problematic. What happens at a threshold of 0.49? Is this considered unsuitable? I am sure that Laabs and Knitter are aware of this methodological pitfall and I would encourage the authors to add some remarks to this section to emphasize the concerns (again, there is reference to table 2, which seems to be substantial to understand the proposed concept).

In the lines 306-321, I am getting a bit lost and at least from my personal perspective it is not easy to follow – maybe also due to some spelling/grammar issues. Because this section seems to be one of the main results, I would consider a minor restructuring and clarification of the findings from the model.

333: You propose up to 17500 farmsteads in the micro-region. Are these numbers based on the assumption of a featureless plain or do they take into account slope gradients, watercourses, soils, and distance to infrastructure etc.? It would be interesting to know what happens if you cluster your farmsteads in a predictive model for rural/hinterland settlements. Would this lead to a reasonable distribution of farms in the complementary region?

In 379 following, you say that the Pergamon area was highly diversified and no regular grid of farmsteads can be assumed. I am sure you were right, but to which extent did you integrate this in your model? Maybe clarify to not create methodological uncertainties in your – as you say in line 392 – simplistic model. Overall, your approach strongly emphasizes this simplicity but it would enhance the paper’s integrity to add some more critical limitations instead of stressing the potential for future applications here…

Eventually, you also stress the vulnerability of the population and particularly the harvest to environmental hazards like droughts (I assume droughts to be the most hazardous events in the region? Maybe flash-floods, plagues?) In a future paper, it would be very interesting to integrate these ‘events’ into the model and simulate different decision-making processes, adaptation strategies, and governmental restrictions etc.

In the individual comments, I raised some questions about the term resilience, which you use to describe the vulnerability of the population to overstraining the landscape

__

General comments

The article is generally well-written, however some minor revision of the English grammar would help to improve the readability (see individual comments section). Please take into account that I am not a native English speaker and I can only provide basic grammatical knowledge. I am sure, this will be solved during the copy-editing process.

Finally, I can recommend this article for publication pending minor revision. In general, the presented approach – although presented as a simple and basic model – is of great interest for future research and I am looking forward to see follow-up articles.

Table 2 was missing in the submission and would have been of significant importance for this review! Please consider this circumstance before evaluating my comments in a broader context.

Individual and minor comments to the authors.

Lines 29 - ..Please include more geographical information in the map, e.g. major hydrologic system (which you mention to delineate the catchments), topographic features like the mountains, and the location of the villages, which you mention in the introduction.

39: ‘defined not primarily geographically’ maybe use ‘by its geographical extent’ because human-environment interaction, which you mention later in the sentence are considered to be equally geographically defined.

50: again it would be nice to see these rivers on the map to help the reader understanding the environmental settings.

53-57: characterized instead of distinguished? Please add a reference to this sentence/section

65: no comma before that, if you want to keep the comma use: which

87: by Becker. Also in this line: Micro-Region instead of micro-region as geographical determination and not Pergamon Micro-Region (name); see line 95.

91: were related

118: something is missing here? … and did not use it his work

126: delete the in the site catchment analysis

126: no comma before that. In general: comma, which, no comma that

126/128: site-catchment or site catchment? Be consistent..

129: please consider that/which instead of what

129: maximum, not maximal

132: comma before which

138: a not as determining factor?

170: built

174: aggregated

170: is 50 by 50 not a bit coarse to simulate landscape accessibility? I am also a bit worried that the TWI is the ultimate tool to assess soil moisture in the study area, particularly when using a 50x50 raster (see Kopecký M., Macek M. & Wild J. (2021) Topographic Wetness Index calculation guidelines based on measured soil moisture and plant species composition. Science of The Total Environment 757: 143785. https://doi.org/10.1016/j.scitotenv.2020.143785 ). However, I am aware that data-availability is an issue in environmental/landscape archaeology and particularly when covering such a large area. Maybe add some critical remarks in the discussion section to emphasize limitations of the method/data availability.

178: I would like to highlight your attempt to the limitations of ‘Landscape reconstructions’ in archaeological research here.

181: It seems like table 2 is missing in the submission? Please provide the table in the revision.

197-199: Please correct me but 81 + 14 + 15 equals 110 and not 100. Also, there is a typo in form/from

201: Pliny? I personally don’t like the translation of names…. Maybe consider Plinius because you also used Strabon (and not the English ‘version’ Strabo).

212: do you mean vulnerable and resilient as opposite of each other? Maybe better use ‘or’ than ‘/’ because (as you know) they have different meanings. (Although the use of the term resilience in archaeology (and in general) needs to be reconsidered due to the limited knowledge that is provided by the archaeological record… but that’s a different story..)

213: missing ‘to’ in fulfill and fulfill missing ‘l’

215: comma, which

247: by Halstead

295: without the missing table 2 it is not easy to follow the argument of the suitability threshold. Although, I would like to believe that the authors derived their results carefully from the calculation.

318: estimated or estimates of?

322: considered as, delete as

328: varies instead of vary

330: ,which

341: suggested by…

354: give hints.. I know what you mean but maybe better: allow conclusions to be drawn, indicate or so?

359: between xx and xxx

363: self-sufficiency, compare self-sufficient and self sufficient in other parts of the manuscript.

364: Pergmene? Please apologize my ignorance but I don’t know what this means, is it a particular term for Pergamon’s chora?

368: were not considered. Check for differences in past/present formulations in the MS

370-375. This sentence is pretty hard to understand due to its length and grammatical structure, please revise.

383: individual farmsteads

384 Consider the Roman Imperial Period and the Hellenistic Period throughout the MS

385: again, Micro-Region or micro-region here?

392: delete of in of our model’s

393: instead of: but the availability of space is. Maybe use: but rather the availability of space

397: this is interesting and deserves some more words and (at least for me) some clarification! Why do you think ‘colonising nature’ decreases resilience? To maintain resilience under growing population pressure and demands, an increase in cropland is the logical consequence to avoid increasing societal vulnerability – please correct me if I am wrong here. You argue that high labour input means transformation of medium or non-suitable land into agricultural cropland, which renders the population very vulnerable.  Also, you say, that with less labour input and increased non-suitable land utilization, the society also operates close to the CC and is hence equally vulnerable to external stressors (crop failure). So, both scenarios generate equal results, which in turn means that the environmental prerequisites did not allow for maximized population numbers proposed from historical sources, right?

Interesting but maybe out of scope of the approach: Is there actually a ‘natural’ perspective in the catchment of Hellenistic/Roman Pergamon and did the people perceive their environment as ‘natural’, which needs to be colonized or is this a mere modern conceptual problem/terminology? The more land gets turned into ‘cultural landscapes’ – whatever that means – the more vulnerable the environment becomes? Like erosion, flooding, droughts? I think this is true but how can we describe this process or measure the actual impact without referring to the ecological concept of resilience (which includes also population development and continuous potential adaptation)?

433: suggests

442: created?

446: focus on?

447: became

451: politically

451: therefore, the relationship…

481, remove comma before does

Author Response

Dear reviewer,

thank you very much for your comments. They were very helpful to improve the paper.

Best wishes,
Julian Laabs & Daniel Knitter

Author Response File: Author Response.pdf